Anatomy and Physiology series: Physiology of Articulation

The physiology of articulation is an incredibly complex act...which is why I felt it necessary to go through the posts on the nervous system first before I attempted to explain any of what we know about it.

To refresh, we're talking about all the muscles of articulation, from the soft palate, to the tongue, to the muscles of the jaw, to the facial muscles.  I thought about counting how many muscles are involved here, but I decided to be lazy and so I didn't...but there are a lot of them.  When you produce a phoneme, or smallest sound in a language (think IPA symbols), your articulators are creating the appropriate, corresponding shape for your vocal tract in relation to that sound.  When you move on to the next sound in a word, the articulators are moving to the position that corresponds to that second sound.  Now add to that the knowledge that we speak an average rate of 10+ phonemes per second (in Standard American English...most other spoken languages are actually faster), and you start to get some idea of how rapidly and precisely these articulatory muscles have to move for someone to produce intelligible, fluent speech.  

Here's the real deal with how we are able to produce such rapid, fine-tuned movement for speech:  We don't really know.  Sigh.  But we do know a few things about how it works:  We use sensory feedback from our muscles and auditory system to learn how to articulate speech in any language.  We also use this sensory feedback to correct our errors in production as well (like if you ever found yourself saying "sable" instead of "table," you probably caught it and corrected yourself).  And motor coordination for the whole act typically happens behind our conscious thought processes in our brain, although it does respond to what we are consciously thinking about (see this post.)

When it comes to the motor-planning aspect of speech, things get a bit muddy.  There are theories that hypothesize that the command for motor-movement comes directly out of linguistic needs; meaning, we coordinate speech in order to communicate something.  Other theories, called dynamic theories, say it's controlled by a series of movements comprising of a system which would work in succession to bring about the articulatory goals.

The current thinking lies somewhere in between those two theory ideas.  The idea is that there are underlying neurological coordinations that control articulation in speech and that there are separate coordinations that control the articulators during non-speech tasks (such as chewing, swallow, and making nonsense sounds).  Much of the evidence for this stems from brain scans and also from the failure of certain therapeutic exercises to work with disordered speech.  See, there used to be a lot of "strength training" exercises for articulation that involved improving muscle tone and flexibility...or so it was thought.  It was believed that these exercises, which are comprised of many, sometimes strange, non-speech tasks and non-speech sounds (some of which crept into the acting and singing world), would help teach children with disordered speech to make the correct sounds.  The failure of these exercises came from two things:  Most disordered speech is not from lack of muscle tone (and ones that are need can't actually do these exercises anyway) and proper execution of non-speech tasks does not generalize to speech-tasks.  So what does that mean?  If a child or adult with disordered speech needs to improve their speech for the purpose of communicating, then they need to practice meaningful speech.*  This is why there's a lot of research going into understanding these coordinations better, if we know what's going on in a normal system then we'll better know how to treat a disordered system.  (We've got some good treatments out there already, but the SLP world is always looking for ways to make their treatments more effective and more efficient...exactly what good voice teachers do in a non-scientific-stuck-in-the-lab-for-years kind of way.)

What does this mean for the singer?  Well, a lot of what I will say here is my hypothesis, but I will say that, based on my own experience with re-learning how to sing as well as experiences teaching these concepts, I think I'm on the right track.  I hypothesize that the articulatory system works best when it is left to function outside conscious thought as much as possible whether speaking or singing (at least in the case of a healthy-functioning system).  Therefore, in order to best train the articulatory system for singing, one needs to train with meaningful speech/communication.  Obviously, vocalizations can be hard to accommodate this, but it seems that when singers think more about communicating something than about making the perfect vowel or sound, then they feel a freedom in the articulatory system automatically.  (And this can work on vocalizations as well:  If I think of making meaningful, musical phrases out of my vocalizations then my articulators don't add unnecessary strain, even when sustaining a single vowel for a long time.)  I believe this is due to the underlying coordination for communication, which would be the built-in, most efficient way to articulate anyway.

How many of us have had the effect of feeling more vocal freedom once we were told to focus on our expression of the text while we sing instead of on how we sound?  And equally, how many of us experienced freedom with our diction in a foreign language after being told to speak it with the proper accent and then sing it?  (And for that matter:  How many variances are there in articulatory movements among the legendary opera singers of the past century?) Although the specific reason for this is still a mystery, I believe the answer lies with how our nervous system already has a built-in system for communicating with our articulators.  We don't usually put our articulators under conscious control in everyday speech, and so if we put them under too much conscious control during singing the system becomes inefficient.  (Like how if you're asked to do someone else's job for a day you'd be much less efficient at that job than that other person.)

That's not to say that they must be under conscious control during some part of our training.  Studies have shown that the brain of professional opera singers do show heightened sensorimotor, prefrontal cortex, and fine motor control activity during singing*; and most training that involves changes to brain activity must be put under conscious control at some point (usually near the beginning of learning a new task).  All I'm saying is our timeline in training is often wrong.  If we train our articulatory system to function efficiently in all the languages we must sing in through speech first, the result is often an efficient system within a few weeks to months, rather than years.  (And if the articulation system isn't free when singing, but is unstrained during speech, then it's usually an issue with compromised laryngeal function rather than the articulatory system itself.)

So up next, I'm going to bring all this A&P stuff together with my current ideas on technique and how current voice science can be applied to learning how to sing.  And if you have any questions, please feel free to post below.  I'll do my best to answer them.  

*Seikel, J. A., King, D. W., & Drumright, D. G. (2010). Anatomy and physiology for speech, language, and hearing. Clifton Park, NY: Delmar.


*Bunton, Kate.  (2008).  Speech versus nonspeech:  Different tasks, different neural organization.  Seminars in Speech and Language, 29(4), 267-275.


*Kleber, B., Veit, R., Birbaumer, N., Gruzelier, J., Lotze, M.  The brain of opera singers: experience-dependent changes in functional activation.  Cerebral Cortex, 20(5), 1144-1152.